- Email: [email protected]
Social interaction and cortisol treatment increase cell addition and radial glia fiber density in the diencephalic periventricular zone of adult electric fish, Apteronotus leptorhynchus
Hormones and Behavior 50 (2006) 10 – 17 www.elsevier.com/locate/yhbeh
Social interaction and cortisol treatment increase cell addition and radial glia fiber density in the diencephalic periventricular zone of adult electric fish, Apteronotus leptorhynchus Kent D. Dunlap ⁎, James F. Castellano 1 , Erealda Prendaj 2 Department of Biology, Trinity College, Hartford, CT 06106, USA Received 22 November 2005; revised 4 January 2006; accepted 6 January 2006 Available online 3 April 2006
Abstract In electric fish, Apteronotus leptorhynchus, both long-term social interaction and cortisol treatment potentiates chirping, an electrocommunication behavior that functions in aggression. Chirping is controlled by the diencephalic prepacemaker nucleus (PPn-C) located just lateral to the ventricle. Cells born in adult proliferative zones such as the periventricular zone (PVZ) can migrate along radial glial fibers to other brain regions, including the PPn-C. We examined whether social interactions or cortisol treatment influenced cell addition and radial glia fiber formation by (1) pairing fish (4 or 7 days) or (2) implanting fish with cortisol (7 or 14 days). Adult fish were injected with bromodeoxyuridine 3 days before sacrifice to mark cells that were recently added. Other fish were sacrificed after 1 or 7 days of treatment to examine vimentin immunoreactivity (IR), a measure of radial glial fiber density. Paired fish had more cell addition than isolated fish at 7 days, coinciding temporally with the onset of socially induced increase in chirping behavior. Paired fish also had higher vimentin IR at 1 and 7 days. For both cell addition and vimentin IR, the effect was regionally specific, increasing in the PVZ adjacent to the PPn-C, but not in surrounding regions. Cortisol increased cell addition at 7 days, correlating with the onset of cortisol-induced changes in chirping, and in a regionally specific manner. Cortisol for 14 days increased cell addition, and cortisol for 7 days increased vimentin IR but in a regionally non-specific manner. The correlation between treatment-induced changes in chirping and regionally specific increases in cell addition, and radial glial fiber formation suggests a causal relationship between such behavioral and brain plasticity in adults, but this hypothesis will require further testing. © 2006 Elsevier Inc. All rights reserved. Keywords: Electrocommunication; Cortisol; BrdU; Radial glia; Vimentin; Cell addition; Glucocorticoids; Electric fish; Apteronotus leptorhynchus; Social behavior
Long-term social interaction and hormone treatment have long been known to influence behavior and its underlying neural processes. Much work has focused on how social environment and hormones achieve their effects on behavior by influencing synaptic plasticity. For example, in rodents, social interaction and glucocorticoid treatment affect learning and simultaneously alter long-term potentiation in the hippocampus (Lu et al., 2003), a brain region that contributes to learning. ⁎ Corresponding author. Fax: +1 860 297 2538. E-mail address: [email protected] (K.D. Dunlap). 1 Present address: Graduate School of Biological Sciences, Mount Sinai School of Medicine, New York, NY 10029, USA. 2 Present address: Department of Neuroscience, University of Connecticut Health Center, Farmington, CT 06030, USA. 0018-506X/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.yhbeh.2006.01.003
In the last decade, researchers have begun to investigate whether social stimuli and hormones affect behavior also by promoting the production and incorporation of new cells into brain regions that control behavior. Cell production in the hippocampus of adult rodents contributes to learning and memory formation (Bruel-Jungerman et al., 2005; Shors et al., 2001, 2002), and such cell production is influenced by the social environment. An enriched environment that includes heightened levels of social interaction dramatically increases cell proliferation in the hippocampus and subventricular zone of rodents (Kempermann and Gage, 1999; Komitova et al., 2005; Kozorovitskiy and Gould, 2004). Steroid treatment also affects cell proliferation in behaviorally important brain regions. For example, cortisol treatment inhibits cell proliferation in the rat hippocampus (Gould et al., 1992).
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
The electrocommunication behavior of weakly electric fish is a useful system for exploring the neural processes underlying socially and hormonally induced behavioral plasticity because of the relatively simple and well-studied link between certain communication behaviors and their control by discrete brain nuclei (Zupanc and Maler, 1997). As one example, Dulka et al. (1995) showed that androgen treatment simultaneously increase the production of an electrocommunication behavior termed chirping and the density of fibers containing substance P, a neuromodulator that regulates chirping, in the brain region that controls chirping. In a previous study, we demonstrated that treatment of electric fish with long-term social interaction or exogenous cortisol potentiated chirping behavior (Dunlap et al., 2002). Here, we examine whether these treatments also affect cell addition and glial fiber formation near the brain region controlling chirping behavior to assess whether behavioral change is associated with structural plasticity in the adult brain. Brown ghost knife fish, Apteronotus leptorhynchus, continually produce high-frequency, quasi-sinusoidal electrical waves termed the electric organ discharge (EOD) that they use for intraspecific communication and for detection of objects in their environment. In certain social contexts, particularly in male–male aggression, they modulate EOD frequency and amplitude to form a communication signal termed a chirp. Chirping is a part of their natural social behavior, but it can also be elicited artificially by stimulating fish with an electrical sine wave at a frequency similar to the fish's own EOD frequency (Dunlap, 2002; Larimer and Macdonald, 1968; Zupanc and Maler, 1993). Our previous study showed that 1 week of social interaction or 1–4 weeks of cortisol treatment increased chirping rate to such standardized stimuli (Dunlap et al., 2002). Chirping is controlled by a diencephalic nucleus, the prepacemaker nucleus (PPn-C), located about 400 μm lateral to the ventricle (for review, see Zupanc and Maler, 1997). It receives input from several regions (e.g., sensory processing regions and the hypothalamus) but projects only to the pacemaker nucleus, which initiates the EOD. Activation of the PPn-C produces a chirp by causing brief (10–100 ms) accelerations of the pacemaker nucleus firing frequency. The singular projection and behavioral output of the PPn-C makes it relatively easy to directly link structural plasticity in the PPn-C with behavioral changes. The brain of adult Apteronotus has an unusually high degree of cell proliferation compared to other vertebrates (Zupanc and Horschke, 1995). Cells are born in many regions of the adult brain but are particularly proliferative in the cerebellum and the periventricular zone (PVZ). The PPn-C also contains a high number of adult-born cells, and, in the closely related genus Eigenmannia, some of the cells born around the ventricular wall migrate laterally to the PPn-C and differentiate into neurons (Zupanc and Zupanc, 1992b). We sought to determine whether cells formed in the ventricular region that give rise to PPn-C neurons are affected by social and hormonal stimuli that also influence chirping behavior. In Apteronotus, much evidence indicates that adult-born cells migrate to brain regions beyond proliferation zones using
11
pathways defined by radial glial fibers (Zupanc and Clint, 2003). Consequently, we also examined whether social stimulation or cortisol treatment influences the density of radial glial fibers and thereby might also enhance the migration of new cells and their incorporation into the PPn-C. Materials and methods Animals Brown ghost knife fish (A. leptorhynchus; Gymnotiformes, Teleosti) were obtained from commercial dealers and housed in individual 38-l tanks that were part of two 1235-l circulating filtration systems. Water temperature was maintained between 26 and 28°C, pH was 6.0–6.5, water conductivity was 500– 600 μs/cm, and the light cycle was 12-h light:12-h dark. Fish were fed frozen brine shrimp or bloodworms every 2 days. All procedures used in this study adhered to ethical standards of animal use specified by the National Institutes of Health (DHEW Publication 80–23). For both sets of experiments – social interaction and cortisol treatment – fish were distributed equally between treatment groups according to body weight because rates of cell proliferation correlate with body size in this species (Zupanc and Horschke, 1995). Fish were also distributed to treatment groups by EOD frequency. EOD frequency is a reliable indicator of sex in this species (Dunlap, 2002; Dunlap et al., 1998), and distributing fish this way ensured that treatment groups were sex balanced. Experimental treatments were very similar to those used in our previous study (Dunlap et al., 2002). For both sets of experimental treatments, we marked recently born cells by injecting fish with bromodeoxyuridine (BrdU, Sigma B-9285), a thymidine analogue incorporated into the DNA during S-phase of the cell cycle. Fish were anesthetized (0.075% 2-phenoxyethanol; Sigma P-1126) and injected i.p. with 30–70 μl solution of BrdU (100 mg/kg) and sacrificed 3 days later. Thus, BrdU marked cells that were born in a ∼4-h period after injection (Zupanc and Horschke, 1995) and that survived the following 3 days. We use the term cell addition to refer to this two part process of cell birth and survival.
Social interaction Fish were divided into two groups, one in which individuals remained isolated and another in which each fish was paired with one other fish of the same sex. We paired fish such that both fish in a pair had similar body size (within 1–2 g) and EOD frequency (within 5–20 Hz). Paired fish were housed in 38-l aquaria partitioned into two equal compartments by a mesh divider that allowed electrical, chemical, and limited visual communication but prevented physical contact. (Fish paired without such a barrier usually kill each other within 24 h.) Isolated fish were housed alone in identical aquaria. We conducted four experiments that differed in the duration of time fish were exposed to social interaction and which antigen was examined. In Experiment 1, fish were paired 1 day before BrdU injection and 4 days before sacrifice (paired fish, n = 4 females and 2 males; isolated fish n = 3 females and 2 males). In Experiment 2, fish were paired 4 days before BrdU injection and 7 days before sacrifice (for both paired and isolated fish, n = 3 females and 2 males). In both experiments, fish were sacrificed 3 days after BrdU injection. In experiments 3 and 4, fish were sacrificed 1 or 7 days after pairing to examine social influences on vimentin expression, as a marker of radial glial fibers (for all treatment groups, n = 3 females and 2 males).
Cortisol treatment Fish were anesthetized as above and implanted with capsules containing cortisol (Sigma H-4001) or empty capsules. Implants consisted of silicone tubes (Konigsberg Instruments) sealed at one end with silicone rubber sealant (Dow Corning) and filled with ∼1–2 mg of cortisol (experimental) or nothing (control). Implants varied in length from 3 to 5 mm, depending on body size of the fish, and were implanted into the peritoneal cavity of anesthetized fish. This
12
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
treatment increases circulating cortisol levels to the upper physiological range in experimental fish while minimally increasing cortisol in control fish (Dunlap et al., 2002). We ran three experiments that differed in treatment duration and which antigen was examined. In Experiment 5, fish were implanted with cortisol capsules (n = 7 females and 5 males) or empty capsules (n = 5 females and 4 males) 4 days before BrdU injection and 7 days before sacrifice. In Experiment 6, fish were implanted with cortisol capsules (n = 2 males and 2 females) or empty capsules (n = 3 females and 3 males) 11 days before BrdU injection and 14 days before sacrifice. In both experiments, fish were sacrificed 3 days after BrdU injection. To test for cortisol effects on radial glial fiber density (Experiment 7), fish were sacrificed 7 days after implantation with cortisol (n = 5 females and 3 males) or empty capsules (n = 4 females and 2 males).
Tissue preparation and immunohistochemistry Brains were fixed in 4% paraformaldehyde (80 min at 4°C), washed in phosphate-buffered solution (0.1 M PBS; 4× 30 min), immersed in 15% sucrose (3 h at 4°C) and 30% sucrose solution (overnight at 4°C). Brains were then rinsed in PBS, frozen in isopentane (30 min at −80°C) mounted in freezing medium and stored −80°C until cryosectioning (40 μm). We used immunohistochemistry with an antibody to BrdU to localize mitotic cells (Experiments 1, 2, 5, 6). To expose the BrdU epitope, sections were first treated with 5 mM citrate buffer (10 min at 90–95°C), washed in PBS (5 min at 37°C), treated with pepsin solution (2.5% pepsin in PBS and 0.1N HCl; 3 min at 37°C), washed in PBS (3 × 5 min at room temperature [RT]), treated with blocking solution (10% donkey serum, 0.3% Triton X in PBS; 1 h at RT), and incubated with sheep anti-BrdU (Capralogics Inc.; 1:100; overnight at 4°C). The sections were washed again with PBS (3 × 10 min), incubated at RT sequentially with avidin and biotin (15 min each), biotin-conjugated donkey anti-sheep IgG (Chemicon International; 1:800; 2 h), and streptavidin conjugated to Alexa 488 (Molecular Probes; 1:800; 1 h). To label radial glial fibers (Experiments 3,4 and 7), sections were blocked (as above), incubated with mouse anti-vimentin (Developmental Studies Hybridoma Bank 1:10, overnight), and donkey anti-mouse IgG conjugated to Cy3 (1:200, Jackson ImmunoResearch Laboratories, Inc.; 2 h at RT). Each run of the immunohistochemistry assay contained matched experimental and control animals of a single experiment.
Statistics We compared density of BrdU+ cells and vimentin IR between sexes and treatment groups using a Mann–Whitney test. There were no significant sex differences in any experiment, and consequently, we pooled data across sexes. Because different axial regions of the PVZ are not independent samples, we compared BrdU+ cell density and vimentin IR between regions within an individual brain using a Wilcoxon sign-rank test for paired samples. All data are presented as mean ± SE, and P b 0.05 was considered statistically significant.
Results Social interaction One day of social interaction (Experiment 1) did not influence rates of cell addition in any region of the diencephalic PVZ (Z = 0.1, P N 0.05, Fig. 1A). Four days of social interaction (Experiment 2) increased cell addition in the PVZ adjacent to the PPn-C (Z = 3.6, P b 0.005) but not in the surrounding regions (Z = 0.7, P N 0.05, Figs. 1B and 2). In paired fish, the density of BrdU+ cells was significantly higher in the PVZ adjacent to the PPn-C than in the PVZ of surrounding regions (Z = 3.0, P b 0.01); there was no significant regional difference in isolated fish (Z = 0.9, P N 0.05). Both 1 day and 7 days of social interaction (Experiments 3 and 4) increased vimentin IR in the PVZ adjacent to the PPn-C
Quantification of immunopositive cells We counted all BrdU-labeled cells and estimated the radial glial fiber density in the periventricular zone (PVZ) of the diencephalon, sections 13–23 of the Apteronotus brain atlas (Maler et al., 1991) using a Nikon Eclipse E600 epifluorescence microscope and a Zeiss LSM 410 confocal microscope. All observers were blind to the experimental treatment when examining the sections. To quantify the abundance of BrdU+ cells, we counted all labeled cells in the PVZ, which we defined as a 100 μm band surrounding the ventricle and which includes both the ventricular and subventricular zone. The BrdU+ cell count for each section was corrected for split cell errors (introduced by counting a cut cell twice in two adjacent sections) using the method described by Zupanc and Horschke (1995). We estimated the PVZ cross-sectional area using NIH Image J 4.0. The volumetric density of BrdU+ cells per section was then calculated as the number of cells in each section divided by PVZ area and section thickness (40 μm). To estimate radial glial fiber density, we measured vimentin immunoreactivity (IR) using NIH Image J 4.0. Confocal images were converted to grayscale, and a similar threshold was set for all images. Mean gray value in an 8-bit image, which ranged from 0 to 255, was measured in the PVZ of all sections. This value correlates positively with the degree of vimentin IR, which is our measure of radial glial fiber density. For all brains, we calculated the mean BrdU+ cell density or glial fiber density in PVZ of 3–6 sections in the region of the diencephalon adjacent to the PPn-C (sections 17–19 in the Apteronotus brain atlas [Maler et al., 1991]), and in the regions just anterior and posterior to the PPn-C (sections 13–16 and 20– 23). Each of these three regions (adjacent, anterior, and posterior to PPn-C) spans an axial distance of about 100 μm.
Fig. 1. Effect of long-term social interaction on cell addition in diencephalic periventricular region of Apteronotus. For definition of the axial regions, see Materials and methods. (A) Four-day treatment (Experiment 1). There were no significant differences in treatment group or in axial region. (B) Seven-day treatment (Experiment 2). Cell addition was significantly elevated only in paired fish in the region adjacent to the prepacemaker nucleus (PPn-C).
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
13
Fig. 2. Sections of the periventricular region adjacent to the prepacemaker nucleus (PPn-C) showing BrdU labeling in fish isolated and paired fish for 7 days (Experiment 2). Arrows indicate positively labeled cells. (A) Isolated fish. (B) Paired fish. (C) Higher magnification of boxed area in 2B. Scale bar = 100 μm.
(Z = 2.3 P b 0.05) but not the surrounding regions (Figs. 3 and 4). In isolated fish in the 1-day treatment (Experiment 3), vimentin IR was significantly lower in the region adjacent to the PPn-C than in the surrounding regions (Fig. 3A; Z = 2.5 P b 0.05). There was no regional difference in vimentin IR in paired fish treated for 1 day or in either treatment group in the 7day treatment (Fig. 3, P N 0.05).
Cortisol treatment Seven days of cortisol treatment (Experiment 5) increased cell addition in the region adjacent to the PPn-C (Z = 2.8, P b 0.05; Fig. 5A) but not in surrounding regions (Z = 0.8, P N 0.05; Fig. 5B). Fourteen days of cortisol treatment (Experiment 6) increased cell addition in both regions of the diencephalon (Z = 1.7, P b 0.05). Fish implanted with cortisol for 7 days (Experiment 5) had significantly higher BrdU+ cell density in the region adjacent to the PPn-C than in surrounding regions; there was no regional difference in isolated fish or in either treatment group in fish treated for 14 days (Experiment 6). Seven day cortisol treatment (Experiment 7) increased vimentin IR in both the region adjacent to the PPn-C (Z = 2.7, P b 0.01) and the surrounding regions (Z = 0.05, P b 0.05; Fig. 6). Discussion
Fig. 3. Effect of long-term social interaction on vimentin immunoreactivity (IR) as a measure of radial glial fiber density in Apteronotus. See Materials and methods for explanation of quantification of IR. (A) One-day treatment (Experiment 3). (B) Seven-day treatment (Experiment 4). For both treatment times, vimentin IR was significantly elevated in paired animals in the region adjacent to the PPn-C, but not in the surrounding region.
We present evidence that both social interaction and cortisol treatment known to potentiate chirping behavior increases cell addition in the PVZ adjacent to the brain nucleus (PPn-C) that controls chirping behavior. These treatments also increase the density of radial glial fibers spanning between the ventricular proliferation zone and the PPn-C and thereby enhance the means by which newly added cells could migrate into the electrocommunication circuitry. To our knowledge, this represents the first demonstration of environmental or hormonal regulation of cell addition or glial fiber density in the adult brain of any fish. At this point, we do not know the phenotype or fate of the new cells stimulated by social and cortisol treatment. Most studies attempting to associate adult cell proliferation with behavioral change have focused on production of neurons (reviewed in Kempermann et al., 2004; Abrous et al., 2005) (However, in a few cases, production of non-neuronal cell has also been associated with behavioral change [Johansson, 2004]). In studies of the closely related electric fish, Eigenmannia, Zupanc identified adult formed cells that
14
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
Fig. 4. Sections of the periventricular region adjacent to the prepacemaker nucleus (PPn-C) showing vimentin labeling in fish isolated or paired for 7 days (Experiment 2). (A) Isolated fish. This section had a vimentin IR score of 12. (B) Paired fish. This section had a vimentin IR of 32. See Fig. 3 for quantification of all treated fish. Scale bar = 100 μm.
differentiate into neurons in the PPn-C, reaching peak abundance with a survival time of 7 days (Zupanc and Zupanc, 1992a), which closely matches the time course of socially and cortisol-induced changes in chirping behavior (Dunlap et al.,
Fig. 6. Effect of 7-day cortisol treatment on vimentin immunoreactivity (IR) in the diencephalic periventricular region of Apteronotus (Experiment 7). Vimentin IR was significantly higher in cortisol-treated fish than in control fish in both regions of the diencephalon.
2002). In other studies on Apteronotus, we found that, after a longer survival time (24 days), about 15% of the BrdU+ cells co-label with Hu, a marker of early neuronal differentiation (data not shown). In zebrafish, about half of the cells born in the adult brain differentiate into neurons after even longer survival times (∼9 months) (Zupanc et al., 2005). Thus, there is abundant evidence that cells formed in the PVZ of adult teleosts, including electric fish, can differentiate into neurons, but more studies are necessary to determine the fate of cells produced in socially and cortisol stimulated Apteronotus over the time course of behavioral change. Social interaction
Fig. 5. Effect of cortisol treatment on cell addition in the diencephalic periventricular region of Apteronotus. (A) Seven-day treatment (Experiment 5). Cell addition was significantly elevated in paired fish in the region adjacent to the prepacemaker nucleus (PPn-C), but not the surrounding regions. (B) Fourteen-day treatment (Experiment 6). Cell addition was significantly elevated in both regions of the diencephalon.
The effect of social interaction on cell addition was regionally and temporally specific. Cell addition and glial fiber formation was stimulated in the PVZ adjacent to the PPn-C but not in the surrounding PVZ. This indicates that the treatment effect is not simply part of a generalized response, such as an overall increase in growth rate or metabolic rate. The onset of socially stimulated cell addition coincided temporally with previously reported changes in chirping behavior (Dunlap et al., 2002) and occurred only after the increased formation of radial glial fibers. This specificity supports the notion that changes in cell addition are somehow causally associated with changes in behavior and glial fiber formation. One possible explanation for the regional and temporal association between cell addition and glial fiber formation is that cell addition may be enhanced selectively in regions that have a suitable migratory environment. In rat development, waves of neurogenesis are often preceded by increased formation of radial glial fibers (Rickmann et al., 1987), and in adult rats, rates of cell proliferation and vimentin expression are closely correlated during several types of experimental manipulations (Alonso, 2001). In another teleost fish, the three-spined stickleback (Gasterosteusa culeatus), regions of highest PVZ cell proliferation occur in so-called migration zones, suggesting a causal relationship between
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
zones of cell proliferation and the presence of migratory substrates such as radial glia (Ekstrom et al., 2001). Our finding that social interaction stimulates cell addition only after enhanced glial fiber formation and only in regions of enhanced glial fiber density supports the idea that glial fiber formation contributes to cell addition. This then raises the question of how social stimulation quickly (within 1 day) increases fiber formation. The regional specificity of socially induced structural changes in the PVZ may arise from social interaction selectively enhancing PVZ cell addition in an activitydependent manner. Certain neurotrophins such as brainderived neurotrophic factor (BDNF) can be secreted in an activity dependent manner (Thomas and Davies, 2005) and can also promote neuron and glial addition (Scharfman et al., 2005; Li et al., 2000). Moreover, enriched environments that include social stimulation simultaneously increase BDNF and neurogenesis in the hippocampus of rats (Young et al., 1999). Thus, in Apteronotus, the PPn-C may increase its production of neurotrophins as it receives more input or produces more output during social interaction, and this increased neurotrophin secretion could then selectively stimulate cell production in the adjacent proliferative region of the PVZ. Alternatively, social stimuli may act on cell addition more indirectly through changing circulating levels of hormones. One week of social interaction increases plasma levels of cortisol (Dunlap et al., 2002), which can independently enhance cell production in the PVZ (see Discussion below). In some birds and mammals, the effect of social interaction on adult brain cell formation often depends on the social status of the subject and the sex of neighboring animals (Fowler et al., 2002; Pravosudov and Omanska, 2005; Kozorovitskiy and Gould, 2004). In our experiment, we could not attribute individual variation in cell addition to the sex of the stimulus fish or to dominance relationships. Fish always received stimulus from a same-sex fish of approximately the same size and EOD frequency. We chose this stimulus because fish are much more likely to communicate using chirps in dyadic interactions when adjacent fish have similar EOD frequencies and because we knew from previous studies that such social stimuli potentiated the chirping response to standardized artificial stimuli (Dunlap et al., 2002). In conducting the experiment this way, we likely avoided setting up a dominance hierarchy since, in this species, dominance is closely related to sex, body size, and EOD frequency (Dunlap, 2002; Dunlap and Oliveri, 2002). Indeed, in almost all cases, both fish of an interacting pair had rates of cell addition that were similarly elevated above the size and sex-matched isolated fish, suggesting that there was no differential response based on social status. Cortisol treatment Seven days of cortisol treatment increased cell addition in a regionally specific manner. Cell addition increased in the region adjacent to the PPn-C, but not in the surrounding region, over a time course that corresponds to the onset of behavioral actions
15
of cortisol. Two weeks of treatment, which also potentiates chirping, had a general stimulatory effect, increasing cell addition both adjacent and distant to the PPn-C. Seven day cortisol treatment also increased radial glial fiber density in the PVZ. This effect of cortisol on glia coincides with the onset of cortisol-induced increases in chirping behavior and cell addition. However, the effect was regionally non-specific. We will need to examine shorter time periods to determine whether a regionally specific effect on glial fiber formation precedes the regionally specific effect on cell addition, as we found with social stimulation. The stimulatory effect of cortisol on cell addition and vimentin IR in electric fish differs from the inhibitory effect widely reported in mammals (Alonso, 2001; Gould et al., 1992, 1997; Fuchs et al., 2001). However, in these mammalian studies, the associated behavior (e.g., learning) is also inhibited by cortisol. Thus, our results are consistent with a positive correlation between behavior and cortisol-induced changes in cell addition and glial fiber formation. Studies in mammals indicate that glucocorticoids affect cell birth in the adult brain by acting directly on glucocorticoid receptors in the cells of proliferative zones (Wong and Herbert, 2005). Glucocorticoid receptor distribution is unknown in Apteronotus, but in another teleost fish (Oncorhynchus mykiss) high density of GR is found in the PVZ of the diencephalon (Teitsma et al., 1998). Cortisol may also stimulate cell proliferation by increasing the production of neurotrophins. Glucocorticoids increase BDNF expression in the rat hippocampus (Chao et al., 1998; Schaaf et al., 1997), and this mechanism has been implicated in glucocorticoid-induced structural plasticity in the hippocampus. In addition, BDNF appears to mediate the effect of testosterone-induced changes in neuron addition in the canary high vocal center, a brain nucleus that is critical for production of song (Rasika et al., 1999). Causal relationship between brain and behavioral plasticity The association between social and cortisol treatment, chirping behavior and structural changes in the diencephalic PVZ raises interesting but, at this point, unanswered questions about causality. The new cells stimulated by social interaction and cortisol treatment could contribute to the accompanying changes in chirping behavior if they become incorporated into or modify the electrocommunication circuitry. Many studies of diverse species have identified correlations between cell formation and behavioral change (Scharff, 2000; Barinaga, 2003), but there is still much debate and little experimental evidence that adult-formed cells can indeed cause changes in the output of the neural circuits that control behavior (reviewed in Kempermann et al., 2004; Abrous et al., 2005). Secondly, changes in behavior induced by these treatments might themselves cause changes in cell addition. Finally, the effects of social interaction and cortisol treatment on chirping behavior and cell addition may be independent of each other. Discerning between these possibilities will require testing for behavioral changes when cell proliferation is disrupted experimentally.
16
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
Acknowledgments The monoclonal antibody to vimentin was developed by A. Alvarez-Buylla and obtained from the Developmental Hybridoma Bank at the University of Iowa. C. Pytte and J. Kirn provided technical and conceptual advice, E. McCarthy collected data for Experiment 7, J. Nord assisted with animal care. V. Salvador, and S. St. Jean assisted with logistics. This work was supported by a grant from NIH (R03-MH066898-01) to KDD and the HHMI to Trinity College. References Abrous, D.N., Koehl, M., Le Moal, M., 2005. Adult neurogenesis: from precursors to network and physiology. Physiol. Rev. 85 (2), 523–569. Alonso, G., 2001. Proliferation of progenitor cells in the adult rat brain correlates with the presence of vimentin-expressing astrocytes. Glia 34 (4), 253–266. Barinaga, M., 2003. Developmental biology. Newborn neurons search for meaning. Science 299 (5603), 32–34. Bruel-Jungerman, E., Laroche, S., Rampon, C., 2005. New neurons in the dentate gyrus are involved in the expression of enhanced long-term memory following environmental enrichment. Eur. J. Neurosci. 21 (2), 513–521. Chao, H.M., Sakai, R.R., Ma, L.Y., McEwen, B.S., 1998. Adrenal steroid regulation of neurotrophic factor expression in the rat hippocampus. Endocrinology 139 (7), 3112–3118. Dulka, J.G., Maler, L., Ellis, W., 1995. Androgen-induced changes in electrocommunicatory behavior are correlated with changes in substance P-like immunoreactivity in the brain of the electric fish Apteronotus leptorhynchus. J. Neurosci. 15 (3), 1879–1890. Dunlap, K.D., 2002. Hormonal and body size correlates of electrocommunication behavior during dyadic interactions in a weakly electric fish, Apteronotus leptorhynchus. Horm. Behav. 41 (2), 187–194. Dunlap, K.D., Oliveri, L.M., 2002. Retreat site selection and social organization in captive electric fish, Apteronotus leptorhynchus. J. Comp. Physiol., A Neuroethol. Sens. Neural Behav. Physiol. 188 (6), 469–477. Dunlap, K.D., Thomas, P., Zakon, H.H., 1998. Diversity of sexual dimorphism in electrocommunication signals and its androgen regulation in a genus of electric fish, Apteronotus. J. Comp. Physiol., A Sens. Neural Behav. Physiol. 183 (1), 77–86. Dunlap, K.D., Pelczar, P.L., Knapp, R., 2002. Social interactions and cortisol treatment increase the production of aggressive electrocommunication signals in male electric fish, Apteronotus leptorhynchus. Horm. Behav. 42 (2), 97–108. Ekstrom, P., Johnsson, C.M., Ohlin, L.M., 2001. Ventricular proliferation zones in the brain of an adult teleost fish and their relation to neuromeres and migration (secondary matrix) zones. J. Comp. Neurol. 436 (1), 92–110. Fuchs, E., Flugge, G., Ohl, F., Lucassen, P., Vollmann-Honsdorf, G.K., Michaelis, T., 2001. Psychosocial stress, glucocorticoids, and structural alterations in the tree shrew hippocampus. Physiol. Behav. 73 (3), 285–291. Fowler, C.D., Liu, Y., Ouimet, C., Wang, Z., 2002. The effects of social environment on adult neurogenesis in the female prairie vole. J. Neurobiol. 51 (2), 115–128. Gould, E., Cameron, H.A., Daniels, D.C., Woolley, C.S., McEwen, B.S., 1992. Adrenal hormones suppress cell division in the adult rat dentate gyrus. J. Neurosci. 12 (9), 3642–3650. Gould, E., McEwen, B.S., Tanapat, P., Galea, L.A., Fuchs, E., 1997. Neurogenesis in the dentate gyrus of the adult tree shrew is regulated by psychosocial stress and NMDA receptor activation. J. Neurosci. 17 (7), 2492–2498. Johansson, B.B., 2004. Functional and cellular effects of environmental enrichment after experimental brain infarcts. Restor. Neurol. Neurosci. 22 (3–5), 163–174. Kempermann, G., Gage, F.H., 1999. Experience-dependent regulation of adult
hippocampal neurogenesis: effects of long-term stimulation and stimulus withdrawal. Hippocampus 9 (3), 321–332. Kempermann, G., Wiskott, L., Gage, F.H., 2004. Functional significance of adult neurogenesis. Curr. Opin. Neurobiol. 14 (2), 186–191. Komitova, M., Mattsson, B., Johansson, B.B., Eriksson, P.S., 2005. Enriched environment increases neural stem/progenitor cell proliferation and neurogenesis in the subventricular zone of stroke-lesioned adult rats. Stroke 36 (6), 1278–1282. Kozorovitskiy, Y., Gould, E., 2004. Dominance hierarchy influences adult neurogenesis in the dentate gyrus. J. Neurosci. 24 (30), 6755–6759. Larimer, J.L., Macdonald, J.A., 1968. Sensory feedback from electroreceptors to electromotor centers in gymnotids. Am. J. Physiol. 214, 1253–1261. Li, X.C., Jarvis, E.D., Alvarez-Borda, B., Lim, D.A., Nottebohm, F., 2000. A relationship between behavior, neurotrophin expression, and new neuron survival. Proc. Natl. Acad. Sci. U. S. A. 97 (15), 8584–8589. Lu, L., Bao, G., Chen, H., Xia, P., Fan, X., Zhang, J., Pei, G., Ma, L., 2003. Modification of hippocampal neurogenesis and neuroplasticity by social environments. Exp. Neurol. 183 (2), 600–669. Maler, L., Sas, E., Johnston, S., Ellis, W., 1991. An atlas of the brain of the electric fish Apteronotus leptorhynchus. J. Chem. Neuroanat. 4, 1–38. Pravosudov, V.V., Omanska, A., 2005. Dominance-related changes in spatial memory are associated with changes in hippocampal cell proliferation rates in mountain chickadees. J. Neurobiol. 62 (1), 31–41. Rasika, S., Alvarez-Buylla, A., Nottebohm, F., 1999. BDNF mediates the effects of testosterone on the survival of new neurons in an adult brain. Neuron 22 (1), 53–62. Rickmann, M., Amaral, D.G., Cowan, W.M., 1987. Organization of radial glial cells during the development of the rat dentate gyrus. J. Comp. Neurol. 264 (4), 449–479. Schaaf, M.J., Hoetelmans, R.W., de Kloet, E.R., Vreugdenhil, E., 1997. Corticosterone regulates expression of BDNF and trkB but not NT-3 and trkC mRNA in the rat hippocampus. J. Neurosci. Res. 48 (4), 334–341. Scharff, C., 2000. Chasing fate and function of new neurons in adult brains. Curr. Opin. Neurobiol. 10 (6), 774–783. Scharfman, H., Goodman, J., Macleod, A., Phani, S., Antonelli, C., Croll, S., 2005. Increased neurogenesis and the ectopic granule cells after intrahippocampal BDNF infusion in adult rats. Exp. Neurol. 192 (2), 348–356. Shors, T.J., Miesegaes, G., Beylin, A., Zhao, M., Rydel, T., Gould, E., 2001. Neurogenesis in the adult is involved in the formation of trace memories. Nature 410 (6826), 372–376. Shors, T.J., Townsend, D.A., Zhao, M., Kozorovitskiy, Y., Gould, E., 2002. Neurogenesis may relate to some but not all types of hippocampaldependent learning. Hippocampus 12 (5), 578–584. Teitsma, C.A., Anglade, I., Toutirais, G., Munoz-Cueto, J.A., Saligaut, D., Ducouret, B., Kah, O., 1998. Immunohistochemical localization of glucocorticoid receptors in the forebrain of the rainbow trout (Oncorhynchus mykiss). J. Comp. Neurol. 401 (3), 395–410. Thomas, K., Davies, A., 2005. Neurotrophins: a ticket to ride for BDNF. Curr. Biol. 15 (7), R262–R264. Wong, E.Y., Herbert, J., 2005. Roles of mineralocorticoid and glucocorticoid receptors in the regulation of progenitor proliferation in the adult hippocampus. Eur. J. Neurosci. 22 (4), 785–792. Young, D., Lawlor, P.A., Leone, P., Dragunow, M., During, M.J., 1999. Environmental enrichment inhibits spontaneous apoptosis, prevents seizures and is neuroprotective. Nat. Med. 5 (4), 448–453. Zupanc, G.K., Clint, S.C., 2003. Potential role of radial glia in adult neurogenesis of teleost fish. Glia 43 (1), 77–86. Zupanc, G.K., Horschke, I., 1995. Proliferation zones in the brain of adult gymnotiform fish: a quantitative mapping study. J. Comp. Neurol. 353 (2), 213–233. Zupanc, G., Maler, L., 1993. Evoked chirping in the weakly electric fish Apteronotus leptorhynchus: a quantitative biophysical analysis. Can. J. Zool. 71, 2301–2310. Zupanc, G.K.H., Maler, L., 1997. Neuronal control of behavioral plasticity: the prepacemaker nucleus of weakly electric gymnotiform fish. J. Comp. Physiol., A Sens. Neural Behav. Physiol. 180, 99–111.
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17 Zupanc, G.K., Zupanc, M.M., 1992a. Birth and migration of neurons in the central posterior/prepacemaker nucleus during adulthood in weakly electric knifefish (Eigenmannia sp.). Proc. Natl. Acad. Sci. U. S. A. 89 (20), 9539–9543. Zupanc, G.K.H., Zupanc, M.M., 1992b. Birth and migration of neurons in the
17
central posterior/prepacemaker nucleus during adulthood in weakly electric knifefish (Eigenmannia sp.). Proc. Natl. Acad. Sci. U. S. A. 89, 9539–9543. Zupanc, G.K., Hinsch, K., Gage, F.H., 2005. Proliferation, migration, neuronal differentiation, and long-term survival of new cells in the adult zebrafish brain. J. Comp. Neurol. 488 (3), 290–319.
Social interaction and cortisol treatment increase cell addition and radial glia fiber density in the diencephalic periventricular zone of adult electric fish, Apteronotus leptorhynchus Kent D. Dunlap ⁎, James F. Castellano 1 , Erealda Prendaj 2 Department of Biology, Trinity College, Hartford, CT 06106, USA Received 22 November 2005; revised 4 January 2006; accepted 6 January 2006 Available online 3 April 2006
Abstract In electric fish, Apteronotus leptorhynchus, both long-term social interaction and cortisol treatment potentiates chirping, an electrocommunication behavior that functions in aggression. Chirping is controlled by the diencephalic prepacemaker nucleus (PPn-C) located just lateral to the ventricle. Cells born in adult proliferative zones such as the periventricular zone (PVZ) can migrate along radial glial fibers to other brain regions, including the PPn-C. We examined whether social interactions or cortisol treatment influenced cell addition and radial glia fiber formation by (1) pairing fish (4 or 7 days) or (2) implanting fish with cortisol (7 or 14 days). Adult fish were injected with bromodeoxyuridine 3 days before sacrifice to mark cells that were recently added. Other fish were sacrificed after 1 or 7 days of treatment to examine vimentin immunoreactivity (IR), a measure of radial glial fiber density. Paired fish had more cell addition than isolated fish at 7 days, coinciding temporally with the onset of socially induced increase in chirping behavior. Paired fish also had higher vimentin IR at 1 and 7 days. For both cell addition and vimentin IR, the effect was regionally specific, increasing in the PVZ adjacent to the PPn-C, but not in surrounding regions. Cortisol increased cell addition at 7 days, correlating with the onset of cortisol-induced changes in chirping, and in a regionally specific manner. Cortisol for 14 days increased cell addition, and cortisol for 7 days increased vimentin IR but in a regionally non-specific manner. The correlation between treatment-induced changes in chirping and regionally specific increases in cell addition, and radial glial fiber formation suggests a causal relationship between such behavioral and brain plasticity in adults, but this hypothesis will require further testing. © 2006 Elsevier Inc. All rights reserved. Keywords: Electrocommunication; Cortisol; BrdU; Radial glia; Vimentin; Cell addition; Glucocorticoids; Electric fish; Apteronotus leptorhynchus; Social behavior
Long-term social interaction and hormone treatment have long been known to influence behavior and its underlying neural processes. Much work has focused on how social environment and hormones achieve their effects on behavior by influencing synaptic plasticity. For example, in rodents, social interaction and glucocorticoid treatment affect learning and simultaneously alter long-term potentiation in the hippocampus (Lu et al., 2003), a brain region that contributes to learning. ⁎ Corresponding author. Fax: +1 860 297 2538. E-mail address: [email protected] (K.D. Dunlap). 1 Present address: Graduate School of Biological Sciences, Mount Sinai School of Medicine, New York, NY 10029, USA. 2 Present address: Department of Neuroscience, University of Connecticut Health Center, Farmington, CT 06030, USA. 0018-506X/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.yhbeh.2006.01.003
In the last decade, researchers have begun to investigate whether social stimuli and hormones affect behavior also by promoting the production and incorporation of new cells into brain regions that control behavior. Cell production in the hippocampus of adult rodents contributes to learning and memory formation (Bruel-Jungerman et al., 2005; Shors et al., 2001, 2002), and such cell production is influenced by the social environment. An enriched environment that includes heightened levels of social interaction dramatically increases cell proliferation in the hippocampus and subventricular zone of rodents (Kempermann and Gage, 1999; Komitova et al., 2005; Kozorovitskiy and Gould, 2004). Steroid treatment also affects cell proliferation in behaviorally important brain regions. For example, cortisol treatment inhibits cell proliferation in the rat hippocampus (Gould et al., 1992).
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
The electrocommunication behavior of weakly electric fish is a useful system for exploring the neural processes underlying socially and hormonally induced behavioral plasticity because of the relatively simple and well-studied link between certain communication behaviors and their control by discrete brain nuclei (Zupanc and Maler, 1997). As one example, Dulka et al. (1995) showed that androgen treatment simultaneously increase the production of an electrocommunication behavior termed chirping and the density of fibers containing substance P, a neuromodulator that regulates chirping, in the brain region that controls chirping. In a previous study, we demonstrated that treatment of electric fish with long-term social interaction or exogenous cortisol potentiated chirping behavior (Dunlap et al., 2002). Here, we examine whether these treatments also affect cell addition and glial fiber formation near the brain region controlling chirping behavior to assess whether behavioral change is associated with structural plasticity in the adult brain. Brown ghost knife fish, Apteronotus leptorhynchus, continually produce high-frequency, quasi-sinusoidal electrical waves termed the electric organ discharge (EOD) that they use for intraspecific communication and for detection of objects in their environment. In certain social contexts, particularly in male–male aggression, they modulate EOD frequency and amplitude to form a communication signal termed a chirp. Chirping is a part of their natural social behavior, but it can also be elicited artificially by stimulating fish with an electrical sine wave at a frequency similar to the fish's own EOD frequency (Dunlap, 2002; Larimer and Macdonald, 1968; Zupanc and Maler, 1993). Our previous study showed that 1 week of social interaction or 1–4 weeks of cortisol treatment increased chirping rate to such standardized stimuli (Dunlap et al., 2002). Chirping is controlled by a diencephalic nucleus, the prepacemaker nucleus (PPn-C), located about 400 μm lateral to the ventricle (for review, see Zupanc and Maler, 1997). It receives input from several regions (e.g., sensory processing regions and the hypothalamus) but projects only to the pacemaker nucleus, which initiates the EOD. Activation of the PPn-C produces a chirp by causing brief (10–100 ms) accelerations of the pacemaker nucleus firing frequency. The singular projection and behavioral output of the PPn-C makes it relatively easy to directly link structural plasticity in the PPn-C with behavioral changes. The brain of adult Apteronotus has an unusually high degree of cell proliferation compared to other vertebrates (Zupanc and Horschke, 1995). Cells are born in many regions of the adult brain but are particularly proliferative in the cerebellum and the periventricular zone (PVZ). The PPn-C also contains a high number of adult-born cells, and, in the closely related genus Eigenmannia, some of the cells born around the ventricular wall migrate laterally to the PPn-C and differentiate into neurons (Zupanc and Zupanc, 1992b). We sought to determine whether cells formed in the ventricular region that give rise to PPn-C neurons are affected by social and hormonal stimuli that also influence chirping behavior. In Apteronotus, much evidence indicates that adult-born cells migrate to brain regions beyond proliferation zones using
11
pathways defined by radial glial fibers (Zupanc and Clint, 2003). Consequently, we also examined whether social stimulation or cortisol treatment influences the density of radial glial fibers and thereby might also enhance the migration of new cells and their incorporation into the PPn-C. Materials and methods Animals Brown ghost knife fish (A. leptorhynchus; Gymnotiformes, Teleosti) were obtained from commercial dealers and housed in individual 38-l tanks that were part of two 1235-l circulating filtration systems. Water temperature was maintained between 26 and 28°C, pH was 6.0–6.5, water conductivity was 500– 600 μs/cm, and the light cycle was 12-h light:12-h dark. Fish were fed frozen brine shrimp or bloodworms every 2 days. All procedures used in this study adhered to ethical standards of animal use specified by the National Institutes of Health (DHEW Publication 80–23). For both sets of experiments – social interaction and cortisol treatment – fish were distributed equally between treatment groups according to body weight because rates of cell proliferation correlate with body size in this species (Zupanc and Horschke, 1995). Fish were also distributed to treatment groups by EOD frequency. EOD frequency is a reliable indicator of sex in this species (Dunlap, 2002; Dunlap et al., 1998), and distributing fish this way ensured that treatment groups were sex balanced. Experimental treatments were very similar to those used in our previous study (Dunlap et al., 2002). For both sets of experimental treatments, we marked recently born cells by injecting fish with bromodeoxyuridine (BrdU, Sigma B-9285), a thymidine analogue incorporated into the DNA during S-phase of the cell cycle. Fish were anesthetized (0.075% 2-phenoxyethanol; Sigma P-1126) and injected i.p. with 30–70 μl solution of BrdU (100 mg/kg) and sacrificed 3 days later. Thus, BrdU marked cells that were born in a ∼4-h period after injection (Zupanc and Horschke, 1995) and that survived the following 3 days. We use the term cell addition to refer to this two part process of cell birth and survival.
Social interaction Fish were divided into two groups, one in which individuals remained isolated and another in which each fish was paired with one other fish of the same sex. We paired fish such that both fish in a pair had similar body size (within 1–2 g) and EOD frequency (within 5–20 Hz). Paired fish were housed in 38-l aquaria partitioned into two equal compartments by a mesh divider that allowed electrical, chemical, and limited visual communication but prevented physical contact. (Fish paired without such a barrier usually kill each other within 24 h.) Isolated fish were housed alone in identical aquaria. We conducted four experiments that differed in the duration of time fish were exposed to social interaction and which antigen was examined. In Experiment 1, fish were paired 1 day before BrdU injection and 4 days before sacrifice (paired fish, n = 4 females and 2 males; isolated fish n = 3 females and 2 males). In Experiment 2, fish were paired 4 days before BrdU injection and 7 days before sacrifice (for both paired and isolated fish, n = 3 females and 2 males). In both experiments, fish were sacrificed 3 days after BrdU injection. In experiments 3 and 4, fish were sacrificed 1 or 7 days after pairing to examine social influences on vimentin expression, as a marker of radial glial fibers (for all treatment groups, n = 3 females and 2 males).
Cortisol treatment Fish were anesthetized as above and implanted with capsules containing cortisol (Sigma H-4001) or empty capsules. Implants consisted of silicone tubes (Konigsberg Instruments) sealed at one end with silicone rubber sealant (Dow Corning) and filled with ∼1–2 mg of cortisol (experimental) or nothing (control). Implants varied in length from 3 to 5 mm, depending on body size of the fish, and were implanted into the peritoneal cavity of anesthetized fish. This
12
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
treatment increases circulating cortisol levels to the upper physiological range in experimental fish while minimally increasing cortisol in control fish (Dunlap et al., 2002). We ran three experiments that differed in treatment duration and which antigen was examined. In Experiment 5, fish were implanted with cortisol capsules (n = 7 females and 5 males) or empty capsules (n = 5 females and 4 males) 4 days before BrdU injection and 7 days before sacrifice. In Experiment 6, fish were implanted with cortisol capsules (n = 2 males and 2 females) or empty capsules (n = 3 females and 3 males) 11 days before BrdU injection and 14 days before sacrifice. In both experiments, fish were sacrificed 3 days after BrdU injection. To test for cortisol effects on radial glial fiber density (Experiment 7), fish were sacrificed 7 days after implantation with cortisol (n = 5 females and 3 males) or empty capsules (n = 4 females and 2 males).
Tissue preparation and immunohistochemistry Brains were fixed in 4% paraformaldehyde (80 min at 4°C), washed in phosphate-buffered solution (0.1 M PBS; 4× 30 min), immersed in 15% sucrose (3 h at 4°C) and 30% sucrose solution (overnight at 4°C). Brains were then rinsed in PBS, frozen in isopentane (30 min at −80°C) mounted in freezing medium and stored −80°C until cryosectioning (40 μm). We used immunohistochemistry with an antibody to BrdU to localize mitotic cells (Experiments 1, 2, 5, 6). To expose the BrdU epitope, sections were first treated with 5 mM citrate buffer (10 min at 90–95°C), washed in PBS (5 min at 37°C), treated with pepsin solution (2.5% pepsin in PBS and 0.1N HCl; 3 min at 37°C), washed in PBS (3 × 5 min at room temperature [RT]), treated with blocking solution (10% donkey serum, 0.3% Triton X in PBS; 1 h at RT), and incubated with sheep anti-BrdU (Capralogics Inc.; 1:100; overnight at 4°C). The sections were washed again with PBS (3 × 10 min), incubated at RT sequentially with avidin and biotin (15 min each), biotin-conjugated donkey anti-sheep IgG (Chemicon International; 1:800; 2 h), and streptavidin conjugated to Alexa 488 (Molecular Probes; 1:800; 1 h). To label radial glial fibers (Experiments 3,4 and 7), sections were blocked (as above), incubated with mouse anti-vimentin (Developmental Studies Hybridoma Bank 1:10, overnight), and donkey anti-mouse IgG conjugated to Cy3 (1:200, Jackson ImmunoResearch Laboratories, Inc.; 2 h at RT). Each run of the immunohistochemistry assay contained matched experimental and control animals of a single experiment.
Statistics We compared density of BrdU+ cells and vimentin IR between sexes and treatment groups using a Mann–Whitney test. There were no significant sex differences in any experiment, and consequently, we pooled data across sexes. Because different axial regions of the PVZ are not independent samples, we compared BrdU+ cell density and vimentin IR between regions within an individual brain using a Wilcoxon sign-rank test for paired samples. All data are presented as mean ± SE, and P b 0.05 was considered statistically significant.
Results Social interaction One day of social interaction (Experiment 1) did not influence rates of cell addition in any region of the diencephalic PVZ (Z = 0.1, P N 0.05, Fig. 1A). Four days of social interaction (Experiment 2) increased cell addition in the PVZ adjacent to the PPn-C (Z = 3.6, P b 0.005) but not in the surrounding regions (Z = 0.7, P N 0.05, Figs. 1B and 2). In paired fish, the density of BrdU+ cells was significantly higher in the PVZ adjacent to the PPn-C than in the PVZ of surrounding regions (Z = 3.0, P b 0.01); there was no significant regional difference in isolated fish (Z = 0.9, P N 0.05). Both 1 day and 7 days of social interaction (Experiments 3 and 4) increased vimentin IR in the PVZ adjacent to the PPn-C
Quantification of immunopositive cells We counted all BrdU-labeled cells and estimated the radial glial fiber density in the periventricular zone (PVZ) of the diencephalon, sections 13–23 of the Apteronotus brain atlas (Maler et al., 1991) using a Nikon Eclipse E600 epifluorescence microscope and a Zeiss LSM 410 confocal microscope. All observers were blind to the experimental treatment when examining the sections. To quantify the abundance of BrdU+ cells, we counted all labeled cells in the PVZ, which we defined as a 100 μm band surrounding the ventricle and which includes both the ventricular and subventricular zone. The BrdU+ cell count for each section was corrected for split cell errors (introduced by counting a cut cell twice in two adjacent sections) using the method described by Zupanc and Horschke (1995). We estimated the PVZ cross-sectional area using NIH Image J 4.0. The volumetric density of BrdU+ cells per section was then calculated as the number of cells in each section divided by PVZ area and section thickness (40 μm). To estimate radial glial fiber density, we measured vimentin immunoreactivity (IR) using NIH Image J 4.0. Confocal images were converted to grayscale, and a similar threshold was set for all images. Mean gray value in an 8-bit image, which ranged from 0 to 255, was measured in the PVZ of all sections. This value correlates positively with the degree of vimentin IR, which is our measure of radial glial fiber density. For all brains, we calculated the mean BrdU+ cell density or glial fiber density in PVZ of 3–6 sections in the region of the diencephalon adjacent to the PPn-C (sections 17–19 in the Apteronotus brain atlas [Maler et al., 1991]), and in the regions just anterior and posterior to the PPn-C (sections 13–16 and 20– 23). Each of these three regions (adjacent, anterior, and posterior to PPn-C) spans an axial distance of about 100 μm.
Fig. 1. Effect of long-term social interaction on cell addition in diencephalic periventricular region of Apteronotus. For definition of the axial regions, see Materials and methods. (A) Four-day treatment (Experiment 1). There were no significant differences in treatment group or in axial region. (B) Seven-day treatment (Experiment 2). Cell addition was significantly elevated only in paired fish in the region adjacent to the prepacemaker nucleus (PPn-C).
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
13
Fig. 2. Sections of the periventricular region adjacent to the prepacemaker nucleus (PPn-C) showing BrdU labeling in fish isolated and paired fish for 7 days (Experiment 2). Arrows indicate positively labeled cells. (A) Isolated fish. (B) Paired fish. (C) Higher magnification of boxed area in 2B. Scale bar = 100 μm.
(Z = 2.3 P b 0.05) but not the surrounding regions (Figs. 3 and 4). In isolated fish in the 1-day treatment (Experiment 3), vimentin IR was significantly lower in the region adjacent to the PPn-C than in the surrounding regions (Fig. 3A; Z = 2.5 P b 0.05). There was no regional difference in vimentin IR in paired fish treated for 1 day or in either treatment group in the 7day treatment (Fig. 3, P N 0.05).
Cortisol treatment Seven days of cortisol treatment (Experiment 5) increased cell addition in the region adjacent to the PPn-C (Z = 2.8, P b 0.05; Fig. 5A) but not in surrounding regions (Z = 0.8, P N 0.05; Fig. 5B). Fourteen days of cortisol treatment (Experiment 6) increased cell addition in both regions of the diencephalon (Z = 1.7, P b 0.05). Fish implanted with cortisol for 7 days (Experiment 5) had significantly higher BrdU+ cell density in the region adjacent to the PPn-C than in surrounding regions; there was no regional difference in isolated fish or in either treatment group in fish treated for 14 days (Experiment 6). Seven day cortisol treatment (Experiment 7) increased vimentin IR in both the region adjacent to the PPn-C (Z = 2.7, P b 0.01) and the surrounding regions (Z = 0.05, P b 0.05; Fig. 6). Discussion
Fig. 3. Effect of long-term social interaction on vimentin immunoreactivity (IR) as a measure of radial glial fiber density in Apteronotus. See Materials and methods for explanation of quantification of IR. (A) One-day treatment (Experiment 3). (B) Seven-day treatment (Experiment 4). For both treatment times, vimentin IR was significantly elevated in paired animals in the region adjacent to the PPn-C, but not in the surrounding region.
We present evidence that both social interaction and cortisol treatment known to potentiate chirping behavior increases cell addition in the PVZ adjacent to the brain nucleus (PPn-C) that controls chirping behavior. These treatments also increase the density of radial glial fibers spanning between the ventricular proliferation zone and the PPn-C and thereby enhance the means by which newly added cells could migrate into the electrocommunication circuitry. To our knowledge, this represents the first demonstration of environmental or hormonal regulation of cell addition or glial fiber density in the adult brain of any fish. At this point, we do not know the phenotype or fate of the new cells stimulated by social and cortisol treatment. Most studies attempting to associate adult cell proliferation with behavioral change have focused on production of neurons (reviewed in Kempermann et al., 2004; Abrous et al., 2005) (However, in a few cases, production of non-neuronal cell has also been associated with behavioral change [Johansson, 2004]). In studies of the closely related electric fish, Eigenmannia, Zupanc identified adult formed cells that
14
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
Fig. 4. Sections of the periventricular region adjacent to the prepacemaker nucleus (PPn-C) showing vimentin labeling in fish isolated or paired for 7 days (Experiment 2). (A) Isolated fish. This section had a vimentin IR score of 12. (B) Paired fish. This section had a vimentin IR of 32. See Fig. 3 for quantification of all treated fish. Scale bar = 100 μm.
differentiate into neurons in the PPn-C, reaching peak abundance with a survival time of 7 days (Zupanc and Zupanc, 1992a), which closely matches the time course of socially and cortisol-induced changes in chirping behavior (Dunlap et al.,
Fig. 6. Effect of 7-day cortisol treatment on vimentin immunoreactivity (IR) in the diencephalic periventricular region of Apteronotus (Experiment 7). Vimentin IR was significantly higher in cortisol-treated fish than in control fish in both regions of the diencephalon.
2002). In other studies on Apteronotus, we found that, after a longer survival time (24 days), about 15% of the BrdU+ cells co-label with Hu, a marker of early neuronal differentiation (data not shown). In zebrafish, about half of the cells born in the adult brain differentiate into neurons after even longer survival times (∼9 months) (Zupanc et al., 2005). Thus, there is abundant evidence that cells formed in the PVZ of adult teleosts, including electric fish, can differentiate into neurons, but more studies are necessary to determine the fate of cells produced in socially and cortisol stimulated Apteronotus over the time course of behavioral change. Social interaction
Fig. 5. Effect of cortisol treatment on cell addition in the diencephalic periventricular region of Apteronotus. (A) Seven-day treatment (Experiment 5). Cell addition was significantly elevated in paired fish in the region adjacent to the prepacemaker nucleus (PPn-C), but not the surrounding regions. (B) Fourteen-day treatment (Experiment 6). Cell addition was significantly elevated in both regions of the diencephalon.
The effect of social interaction on cell addition was regionally and temporally specific. Cell addition and glial fiber formation was stimulated in the PVZ adjacent to the PPn-C but not in the surrounding PVZ. This indicates that the treatment effect is not simply part of a generalized response, such as an overall increase in growth rate or metabolic rate. The onset of socially stimulated cell addition coincided temporally with previously reported changes in chirping behavior (Dunlap et al., 2002) and occurred only after the increased formation of radial glial fibers. This specificity supports the notion that changes in cell addition are somehow causally associated with changes in behavior and glial fiber formation. One possible explanation for the regional and temporal association between cell addition and glial fiber formation is that cell addition may be enhanced selectively in regions that have a suitable migratory environment. In rat development, waves of neurogenesis are often preceded by increased formation of radial glial fibers (Rickmann et al., 1987), and in adult rats, rates of cell proliferation and vimentin expression are closely correlated during several types of experimental manipulations (Alonso, 2001). In another teleost fish, the three-spined stickleback (Gasterosteusa culeatus), regions of highest PVZ cell proliferation occur in so-called migration zones, suggesting a causal relationship between
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
zones of cell proliferation and the presence of migratory substrates such as radial glia (Ekstrom et al., 2001). Our finding that social interaction stimulates cell addition only after enhanced glial fiber formation and only in regions of enhanced glial fiber density supports the idea that glial fiber formation contributes to cell addition. This then raises the question of how social stimulation quickly (within 1 day) increases fiber formation. The regional specificity of socially induced structural changes in the PVZ may arise from social interaction selectively enhancing PVZ cell addition in an activitydependent manner. Certain neurotrophins such as brainderived neurotrophic factor (BDNF) can be secreted in an activity dependent manner (Thomas and Davies, 2005) and can also promote neuron and glial addition (Scharfman et al., 2005; Li et al., 2000). Moreover, enriched environments that include social stimulation simultaneously increase BDNF and neurogenesis in the hippocampus of rats (Young et al., 1999). Thus, in Apteronotus, the PPn-C may increase its production of neurotrophins as it receives more input or produces more output during social interaction, and this increased neurotrophin secretion could then selectively stimulate cell production in the adjacent proliferative region of the PVZ. Alternatively, social stimuli may act on cell addition more indirectly through changing circulating levels of hormones. One week of social interaction increases plasma levels of cortisol (Dunlap et al., 2002), which can independently enhance cell production in the PVZ (see Discussion below). In some birds and mammals, the effect of social interaction on adult brain cell formation often depends on the social status of the subject and the sex of neighboring animals (Fowler et al., 2002; Pravosudov and Omanska, 2005; Kozorovitskiy and Gould, 2004). In our experiment, we could not attribute individual variation in cell addition to the sex of the stimulus fish or to dominance relationships. Fish always received stimulus from a same-sex fish of approximately the same size and EOD frequency. We chose this stimulus because fish are much more likely to communicate using chirps in dyadic interactions when adjacent fish have similar EOD frequencies and because we knew from previous studies that such social stimuli potentiated the chirping response to standardized artificial stimuli (Dunlap et al., 2002). In conducting the experiment this way, we likely avoided setting up a dominance hierarchy since, in this species, dominance is closely related to sex, body size, and EOD frequency (Dunlap, 2002; Dunlap and Oliveri, 2002). Indeed, in almost all cases, both fish of an interacting pair had rates of cell addition that were similarly elevated above the size and sex-matched isolated fish, suggesting that there was no differential response based on social status. Cortisol treatment Seven days of cortisol treatment increased cell addition in a regionally specific manner. Cell addition increased in the region adjacent to the PPn-C, but not in the surrounding region, over a time course that corresponds to the onset of behavioral actions
15
of cortisol. Two weeks of treatment, which also potentiates chirping, had a general stimulatory effect, increasing cell addition both adjacent and distant to the PPn-C. Seven day cortisol treatment also increased radial glial fiber density in the PVZ. This effect of cortisol on glia coincides with the onset of cortisol-induced increases in chirping behavior and cell addition. However, the effect was regionally non-specific. We will need to examine shorter time periods to determine whether a regionally specific effect on glial fiber formation precedes the regionally specific effect on cell addition, as we found with social stimulation. The stimulatory effect of cortisol on cell addition and vimentin IR in electric fish differs from the inhibitory effect widely reported in mammals (Alonso, 2001; Gould et al., 1992, 1997; Fuchs et al., 2001). However, in these mammalian studies, the associated behavior (e.g., learning) is also inhibited by cortisol. Thus, our results are consistent with a positive correlation between behavior and cortisol-induced changes in cell addition and glial fiber formation. Studies in mammals indicate that glucocorticoids affect cell birth in the adult brain by acting directly on glucocorticoid receptors in the cells of proliferative zones (Wong and Herbert, 2005). Glucocorticoid receptor distribution is unknown in Apteronotus, but in another teleost fish (Oncorhynchus mykiss) high density of GR is found in the PVZ of the diencephalon (Teitsma et al., 1998). Cortisol may also stimulate cell proliferation by increasing the production of neurotrophins. Glucocorticoids increase BDNF expression in the rat hippocampus (Chao et al., 1998; Schaaf et al., 1997), and this mechanism has been implicated in glucocorticoid-induced structural plasticity in the hippocampus. In addition, BDNF appears to mediate the effect of testosterone-induced changes in neuron addition in the canary high vocal center, a brain nucleus that is critical for production of song (Rasika et al., 1999). Causal relationship between brain and behavioral plasticity The association between social and cortisol treatment, chirping behavior and structural changes in the diencephalic PVZ raises interesting but, at this point, unanswered questions about causality. The new cells stimulated by social interaction and cortisol treatment could contribute to the accompanying changes in chirping behavior if they become incorporated into or modify the electrocommunication circuitry. Many studies of diverse species have identified correlations between cell formation and behavioral change (Scharff, 2000; Barinaga, 2003), but there is still much debate and little experimental evidence that adult-formed cells can indeed cause changes in the output of the neural circuits that control behavior (reviewed in Kempermann et al., 2004; Abrous et al., 2005). Secondly, changes in behavior induced by these treatments might themselves cause changes in cell addition. Finally, the effects of social interaction and cortisol treatment on chirping behavior and cell addition may be independent of each other. Discerning between these possibilities will require testing for behavioral changes when cell proliferation is disrupted experimentally.
16
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17
Acknowledgments The monoclonal antibody to vimentin was developed by A. Alvarez-Buylla and obtained from the Developmental Hybridoma Bank at the University of Iowa. C. Pytte and J. Kirn provided technical and conceptual advice, E. McCarthy collected data for Experiment 7, J. Nord assisted with animal care. V. Salvador, and S. St. Jean assisted with logistics. This work was supported by a grant from NIH (R03-MH066898-01) to KDD and the HHMI to Trinity College. References Abrous, D.N., Koehl, M., Le Moal, M., 2005. Adult neurogenesis: from precursors to network and physiology. Physiol. Rev. 85 (2), 523–569. Alonso, G., 2001. Proliferation of progenitor cells in the adult rat brain correlates with the presence of vimentin-expressing astrocytes. Glia 34 (4), 253–266. Barinaga, M., 2003. Developmental biology. Newborn neurons search for meaning. Science 299 (5603), 32–34. Bruel-Jungerman, E., Laroche, S., Rampon, C., 2005. New neurons in the dentate gyrus are involved in the expression of enhanced long-term memory following environmental enrichment. Eur. J. Neurosci. 21 (2), 513–521. Chao, H.M., Sakai, R.R., Ma, L.Y., McEwen, B.S., 1998. Adrenal steroid regulation of neurotrophic factor expression in the rat hippocampus. Endocrinology 139 (7), 3112–3118. Dulka, J.G., Maler, L., Ellis, W., 1995. Androgen-induced changes in electrocommunicatory behavior are correlated with changes in substance P-like immunoreactivity in the brain of the electric fish Apteronotus leptorhynchus. J. Neurosci. 15 (3), 1879–1890. Dunlap, K.D., 2002. Hormonal and body size correlates of electrocommunication behavior during dyadic interactions in a weakly electric fish, Apteronotus leptorhynchus. Horm. Behav. 41 (2), 187–194. Dunlap, K.D., Oliveri, L.M., 2002. Retreat site selection and social organization in captive electric fish, Apteronotus leptorhynchus. J. Comp. Physiol., A Neuroethol. Sens. Neural Behav. Physiol. 188 (6), 469–477. Dunlap, K.D., Thomas, P., Zakon, H.H., 1998. Diversity of sexual dimorphism in electrocommunication signals and its androgen regulation in a genus of electric fish, Apteronotus. J. Comp. Physiol., A Sens. Neural Behav. Physiol. 183 (1), 77–86. Dunlap, K.D., Pelczar, P.L., Knapp, R., 2002. Social interactions and cortisol treatment increase the production of aggressive electrocommunication signals in male electric fish, Apteronotus leptorhynchus. Horm. Behav. 42 (2), 97–108. Ekstrom, P., Johnsson, C.M., Ohlin, L.M., 2001. Ventricular proliferation zones in the brain of an adult teleost fish and their relation to neuromeres and migration (secondary matrix) zones. J. Comp. Neurol. 436 (1), 92–110. Fuchs, E., Flugge, G., Ohl, F., Lucassen, P., Vollmann-Honsdorf, G.K., Michaelis, T., 2001. Psychosocial stress, glucocorticoids, and structural alterations in the tree shrew hippocampus. Physiol. Behav. 73 (3), 285–291. Fowler, C.D., Liu, Y., Ouimet, C., Wang, Z., 2002. The effects of social environment on adult neurogenesis in the female prairie vole. J. Neurobiol. 51 (2), 115–128. Gould, E., Cameron, H.A., Daniels, D.C., Woolley, C.S., McEwen, B.S., 1992. Adrenal hormones suppress cell division in the adult rat dentate gyrus. J. Neurosci. 12 (9), 3642–3650. Gould, E., McEwen, B.S., Tanapat, P., Galea, L.A., Fuchs, E., 1997. Neurogenesis in the dentate gyrus of the adult tree shrew is regulated by psychosocial stress and NMDA receptor activation. J. Neurosci. 17 (7), 2492–2498. Johansson, B.B., 2004. Functional and cellular effects of environmental enrichment after experimental brain infarcts. Restor. Neurol. Neurosci. 22 (3–5), 163–174. Kempermann, G., Gage, F.H., 1999. Experience-dependent regulation of adult
hippocampal neurogenesis: effects of long-term stimulation and stimulus withdrawal. Hippocampus 9 (3), 321–332. Kempermann, G., Wiskott, L., Gage, F.H., 2004. Functional significance of adult neurogenesis. Curr. Opin. Neurobiol. 14 (2), 186–191. Komitova, M., Mattsson, B., Johansson, B.B., Eriksson, P.S., 2005. Enriched environment increases neural stem/progenitor cell proliferation and neurogenesis in the subventricular zone of stroke-lesioned adult rats. Stroke 36 (6), 1278–1282. Kozorovitskiy, Y., Gould, E., 2004. Dominance hierarchy influences adult neurogenesis in the dentate gyrus. J. Neurosci. 24 (30), 6755–6759. Larimer, J.L., Macdonald, J.A., 1968. Sensory feedback from electroreceptors to electromotor centers in gymnotids. Am. J. Physiol. 214, 1253–1261. Li, X.C., Jarvis, E.D., Alvarez-Borda, B., Lim, D.A., Nottebohm, F., 2000. A relationship between behavior, neurotrophin expression, and new neuron survival. Proc. Natl. Acad. Sci. U. S. A. 97 (15), 8584–8589. Lu, L., Bao, G., Chen, H., Xia, P., Fan, X., Zhang, J., Pei, G., Ma, L., 2003. Modification of hippocampal neurogenesis and neuroplasticity by social environments. Exp. Neurol. 183 (2), 600–669. Maler, L., Sas, E., Johnston, S., Ellis, W., 1991. An atlas of the brain of the electric fish Apteronotus leptorhynchus. J. Chem. Neuroanat. 4, 1–38. Pravosudov, V.V., Omanska, A., 2005. Dominance-related changes in spatial memory are associated with changes in hippocampal cell proliferation rates in mountain chickadees. J. Neurobiol. 62 (1), 31–41. Rasika, S., Alvarez-Buylla, A., Nottebohm, F., 1999. BDNF mediates the effects of testosterone on the survival of new neurons in an adult brain. Neuron 22 (1), 53–62. Rickmann, M., Amaral, D.G., Cowan, W.M., 1987. Organization of radial glial cells during the development of the rat dentate gyrus. J. Comp. Neurol. 264 (4), 449–479. Schaaf, M.J., Hoetelmans, R.W., de Kloet, E.R., Vreugdenhil, E., 1997. Corticosterone regulates expression of BDNF and trkB but not NT-3 and trkC mRNA in the rat hippocampus. J. Neurosci. Res. 48 (4), 334–341. Scharff, C., 2000. Chasing fate and function of new neurons in adult brains. Curr. Opin. Neurobiol. 10 (6), 774–783. Scharfman, H., Goodman, J., Macleod, A., Phani, S., Antonelli, C., Croll, S., 2005. Increased neurogenesis and the ectopic granule cells after intrahippocampal BDNF infusion in adult rats. Exp. Neurol. 192 (2), 348–356. Shors, T.J., Miesegaes, G., Beylin, A., Zhao, M., Rydel, T., Gould, E., 2001. Neurogenesis in the adult is involved in the formation of trace memories. Nature 410 (6826), 372–376. Shors, T.J., Townsend, D.A., Zhao, M., Kozorovitskiy, Y., Gould, E., 2002. Neurogenesis may relate to some but not all types of hippocampaldependent learning. Hippocampus 12 (5), 578–584. Teitsma, C.A., Anglade, I., Toutirais, G., Munoz-Cueto, J.A., Saligaut, D., Ducouret, B., Kah, O., 1998. Immunohistochemical localization of glucocorticoid receptors in the forebrain of the rainbow trout (Oncorhynchus mykiss). J. Comp. Neurol. 401 (3), 395–410. Thomas, K., Davies, A., 2005. Neurotrophins: a ticket to ride for BDNF. Curr. Biol. 15 (7), R262–R264. Wong, E.Y., Herbert, J., 2005. Roles of mineralocorticoid and glucocorticoid receptors in the regulation of progenitor proliferation in the adult hippocampus. Eur. J. Neurosci. 22 (4), 785–792. Young, D., Lawlor, P.A., Leone, P., Dragunow, M., During, M.J., 1999. Environmental enrichment inhibits spontaneous apoptosis, prevents seizures and is neuroprotective. Nat. Med. 5 (4), 448–453. Zupanc, G.K., Clint, S.C., 2003. Potential role of radial glia in adult neurogenesis of teleost fish. Glia 43 (1), 77–86. Zupanc, G.K., Horschke, I., 1995. Proliferation zones in the brain of adult gymnotiform fish: a quantitative mapping study. J. Comp. Neurol. 353 (2), 213–233. Zupanc, G., Maler, L., 1993. Evoked chirping in the weakly electric fish Apteronotus leptorhynchus: a quantitative biophysical analysis. Can. J. Zool. 71, 2301–2310. Zupanc, G.K.H., Maler, L., 1997. Neuronal control of behavioral plasticity: the prepacemaker nucleus of weakly electric gymnotiform fish. J. Comp. Physiol., A Sens. Neural Behav. Physiol. 180, 99–111.
K.D. Dunlap et al. / Hormones and Behavior 50 (2006) 10–17 Zupanc, G.K., Zupanc, M.M., 1992a. Birth and migration of neurons in the central posterior/prepacemaker nucleus during adulthood in weakly electric knifefish (Eigenmannia sp.). Proc. Natl. Acad. Sci. U. S. A. 89 (20), 9539–9543. Zupanc, G.K.H., Zupanc, M.M., 1992b. Birth and migration of neurons in the
17
central posterior/prepacemaker nucleus during adulthood in weakly electric knifefish (Eigenmannia sp.). Proc. Natl. Acad. Sci. U. S. A. 89, 9539–9543. Zupanc, G.K., Hinsch, K., Gage, F.H., 2005. Proliferation, migration, neuronal differentiation, and long-term survival of new cells in the adult zebrafish brain. J. Comp. Neurol. 488 (3), 290–319.